SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?


To: kapkan4u who wrote (47194)7/12/2001 1:28:35 AM
From: kapkan4uRespond to of 275872
 
theinquirer.net

A quote: "Any such move would have meant a considerable commitment for both companies, and AMD, at a very senior level in the company, felt it had insufficient resources to spread its activities wider, the INQUIRER has discovered. Instead, it will continue developing its X86-64 platform as an alternative to the Itanium platform."

Early, when first x86-64 specs were being kicked around AMD there were some who were pushing to implement 64-bit Alpha ISA instead. AMD would have been in deep sh*t today with such x86-32/Alpha-64 CPU because of Compaq's pulling the plug on Alpha.

Kap



To: kapkan4u who wrote (47194)7/12/2001 8:44:33 AM
From: that_crazy_dougRespond to of 275872
 
<< The scariest thing for AMD that Intel could do is to junk P4 for Tualatin and junk Itanic for x86-64 extended Tualatin. >>

I don't think this would be scary at all, in fact I'd buy all my AMD back if this happened. Tualatin on .13 seems to be less then Athlon at .18 performance and mhz wise, I think this is the main reason we won't see this happen. The performance of the best tualatins is probably about the same as the best p4s, but the p4s certainly have the mhz press advantage, and when the p4s get shrunk they'll most certainly smoke the tualatins in performance as well.

<< Minor changes to PIII's pipeline could keep the MHz in P4's ballpark with much better IPC, smaller die and cheaper platform. >>

I don't think there's any such minor change available to be made. I think you greatly simplify the amount of work required to extend the pipeline. The amount of work here is probably several years to get into production.



To: kapkan4u who wrote (47194)7/12/2001 11:17:23 AM
From: dale_laroyRespond to of 275872
 
>The scariest thing for AMD that Intel could do is to junk P4 for Tualatin and junk Itanic for x86-64 extended Tualatin.<

Junking Itanic for x86-64 would be a major mistake. Imagine how vendors, such as Compaq, Dell, HP, IBM, and Microsoft would feel towards Intel after all they have invested in Itanic if Intel were to simply junk it.

Nevertheless, preannouncing a new x86-64 architecture could inflict heavy damage to AMD. Intel could preannounce a new x86-64 architecture that would be closer to what AMD's original architecture than what the Hammer series is now slated to be. By this I mean that Intel could promise that it would feature triadic operators in x86-64 mode for both integer and floating-point code.

>Minor changes to PIII's pipeline could keep the MHz in P4's ballpark with much better IPC, smaller die and cheaper platform.<

The only minor change likely to keep the PIII's MHz in the P4 ballpark would be doubling the clock and allocating two clocks for the completion of every operation that currently takes one clock. This would produce a 2.0 GHz processor with identical performance to a 1.0 GHz PIII.